Despite the knowledge from randomized controlled clinical trials for the efficacy of medical treatments for CHF, effectiveness of drugs in everyday use is not well-documented. In particular, although there is substantial evidence for the use of beta blockers and angiotensin-converting enzyme (ACE) inhibitors for the treatment of heart failure, there is less conclusive evidence for the use of angiotensin receptor blockers (ARBs), and there is a lack of clear evidence for the effectiveness of these drugs in elderly populations. Finally, there is almost no evidence for the comparative effectiveness of individual drugs within therapeutic class. The purpose of this project is to examine the outcomes of treatment with angiotensin receptor blockers in a national cohort of elderly patients with CHF. Specific Aim #1: To use Cox proportional hazards regression to measure the effect of angiotensin receptor blocker use on hospitalization and mortality. Specific Aim #2: To compare results from Aim 1 to estimates obtained from marginal structural models. Data from a national cohort of patients with heart failure from the Department of Veterans Affairs (VA) will be used in this retrospective cohort study. VA prescription fill records from 1999 through 2003 have been linked to VA inpatient and outpatient utilization and mortality records. These data will be cross-linked to Medicare files to improve ascertainment of the primary outcomes of hospitalization and mortality. Marginal structural models using inverse probability of treatment weighting will be used to test treatment effects, and results will be compared to results obtained from Cox proportional hazards regression. The project will obtain empirical results of outcomes and effectiveness of ARBs for treatment of heart failure, a highly prevalent condition with substantial morbidity and mortality in elderly patients. The methods to strengthen causal inference for observational studies are relatively new and will improve the validity of findings. With the escalation in drug costs, aging of the U.S. population, and extension of drug coverage to the over 42 million Medicare beneficiaries, many of whom have heart failure, more information about outcomes and comparative effectiveness of drugs in everyday use is critically needed. Project Narrative Although there is good evidence for the treatment of chronic heart failure, there is a lack of consensus regarding the efficacy of some drugs in the elderly, because older patients are not included as often in the clinical trials that approve the drugs. The current study will examine the outcomes of treatment with angiotensin II receptor blockers (ARBs), a relatively new class of drugs for the treatment of heart failure, to see which are most effective in terms of reducing hospitalization and improving survival. [unreadable] [unreadable] [unreadable]